Stereo photogrammetry is a fundamental technique for 3D measurements in different applications. As passive stereo methods struggle in texture-less regions, different light-coded patterns are projected to solve the correspondence problem. When the reconstruction method requires a series of sequentially coded light patterns, the system is prone to movement-related errors. A single pattern is utilized where the potential subjects are dynamic or change the position rapidly. A random speckle pattern is a good candidate for such systems. Primarily, there are two approaches to generate the speckle pattern in stereoscopic systems. The speckles can be generated by the coherent illumination of a laser diode, and the laser speckle-like pattern can also be simulated and projected by a professional projector. The purpose of both is to solve the stereo correspondence problem; however, the performance of both can differ, subjective to employed 3D sensing algorithm. In this Letter, we compare the performance of both for single-shot 3D reconstruction. We have discussed the pros and cons of both methods and analyze their ability to resolve long range 3D reconstruction.